PHYSICS. 265 



Gouy has investigated the photometry of colored flames, and has de- 

 vised a spectroscope by which various sources of light of diftereut colors 

 may be compared with others, the comparison being made in parts of 

 their spectra having the same tints. The spectroscope is an ordinary 

 two-prism instrument, the collimator having its objective half covered 

 with a plate of plane glass so iucliued that the light from a second colli- 

 mator falling on its face is reflected into the i)risms. The eye, therefore, 

 sees in the observing telescope two spectra in exact coincidence, one 

 coming from each of the two collimators. One of these comes from a 

 ■gas flame, and is continuous ; the other is a bright line spectrum com- 

 ing from an incandescent vapor. By means of two Nicol prisms the 

 intensity of the former may be regulated. The angle between these 

 prisms measiu'es the illuminating i)Ower. The author shows that for 

 wide bands, and for contiiuious spectra, their transi)arence is complete. 

 For narrow lines, it appears not to be true; that for each simple ray the 

 emissive and the absorptive power depend only on the quantity of vapor 

 contained in a right cylinder having a unit base, and for its height the 

 thickness of the layer; that two layers of vapor at the same temperature, 

 which give the same brilliance to a given line, have identically the same 

 vapor quantity aiul are optically alike; and that the widening of lines 

 is a necessary consequence of the fact that the radiation of the vapor is 

 a continuous function of the wave-length, which does not disap])ear be- 

 yond the lines, but fornu, a background of continuous s])ectrum, which 

 is never wanting when the illumination is sufiicieutly brilliant. — {Ann. 

 Ghim. Phys., V, xviii,5, September, 1879; J. Plujs., ix, 19, January, 1880.) 



Michelson read at the Saratoga meeting of the American Association 

 a pajjer describing his experimental determination of the velocity of light. 

 The method which he employed was Foucault's, with a revolving mirror, 

 but in which some modifications were introduced. The chief of these 

 was the replacement of Foucault's si)herical mirror by a plane mirror, 

 thus enabling the distance between this mirror and the revolving mirror 

 to be increased indefinitely. To make the displacement of the returned 

 image as great as possible, the distance from the slit to the revolving 

 mirror and the speed of rotation should also be increased. But increas- 

 ing the first distance diminishes the second, from the laws of conjugate 

 foci. This difficulty was overcome by using a lens of 150 feet focus, and 

 by placing the revolving mirror 15 feet within the principal focus of the 

 lens. In the experiments the distance between the two mirrors was 

 nearly 2,000 feet, and the distance from the slit to the revolving mirror 

 was about 30 feet, the speed of the rotating mirror being about 257 rev- 

 olutions per second. The deflection obtained exceeded 133™"*, being 

 about 200 times that obtained by Foucault. This deflection was meas- 

 ured to within three or four hundredths of a millimeter, and the result 

 therefore may be regarded as correct within one ten-thousandth part. 

 The mirror was rotated by an air turbine on the same axle, the i)ressure 

 of air being kept constant by means of a water-gauge. An electrically 



